2 papers
cs.AI2026
Reason Less, Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents
Vikas Reddy, Sumanth Reddy Challaram, Abhishek Basu
Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully. In policy-permissive environments, a tool may ex…
cs.AI2026
Reliable Post-Retrieval Assembly for Agent Memory: Separating Evidence Extraction from Policy Execution
Vikas Reddy, Sumanth Challaram, Sumanth Reddy Challaram
LLM-based memory systems can retrieve relevant evidence yet still fail when answer generation entangles semantic filtering, conflict resolution, prior suppression, and output gener…